1 Introduction^*

The tense-form ιαται in Mark 5:29 is universally accepted by interpreters as a perfect passive third person singular form of ἰάομαι: καὶ εὐθὺς ἐξηράνθη ἡ πηγὴ τοῦ αἵματος αὐτῆς καὶ ἔγνω τῷ σώματι ὅτι ἴαται ἀπὸ τῆς μάστιγος (NA²⁸).¹ In this short study, I argue that ιαται in Mark 5:29 should not be read as ἴαται (perfect tense-form) but ἰᾶται (present tense-form) based on the manuscript evidence, the scarcity of the perfect indicative third person singular of ἰάομαι in extant literature, and Mark’s stylistic tendency to render indirect internal discourse using a concatenation of elements including verbs in the perfect tense-form. This reading is confirmed by some of the witnesses from the Old Latin tradition. I then interrogate the lexical semantics of ἰᾶται in ancient Greek literature to confirm the typical understanding of the verb as functioning as a passive middle. Finally, I re-assess the grammatical subject of the verb suggesting that it is not the woman herself who is the grammatical subject (“she”) but her body (“it”).

2 Accenting ιαται in Mark 5:29

Critical editions of the Greek New Testament (GNT), like the NA²⁸, provide accents not present in our earliest manuscripts. In our earliest witnesses of Mark, such as Codex Sinaiticus (01) or Codex Vaticanus (03), scribes only included the letters ΙΑΤΑΙ without accents, which morphologically could refer either to the perfect indicative third person singular form (ἴαται) or to the present indicative third person singular form (ἰᾶται). New Testament manuscripts before the 7th century did not uniformly include accents which clearly differentiated tense.² However, among the later witnesses that do have accents, the perfect accentuation of ιαται dominates, especially among uncials. Majuscules that accent ιαται as a perfect include Codex Basilensis (07), Codex Boreelianus (09), Codex Seidelianus II (013), Codex Vaticanus 354 (028) and Codex Sangallensis (037). The majority of 9th century minuscules accent ιαται as a perfect (e.g. 399 566 892 1424 2811 l 63 l 848 l 849), a trend which continues. Following the majority of majuscules and minuscules that preserve some kind of accentuation, the verb ιαται in Mark 5:29 has been assigned the accents for the perfect indicative form of ἰάομαι in all major editions of the GNT (e.g. Bengel, Alter, Griesbach, Soden, Tischendorf, Tregelles, Westcott and Hort, Pierpont/Robinson and most recently the Tyndale House GNT).³

To my knowledge, the only edition of the GNT that accents ιαται in Mark 5:29 with the present tense-form is Erasmus’ Novum (Instrumentum) Testamentum omne.⁴ This is likely due to the limited manuscripts he had at his disposal. His decision to render ιαται as ἰᾶται instead of ἴαται might be explained by the simple fact that out of the two minuscules of the Gospels we know he used for his first edition, 2 and 817, he favoured 817 which happens to have the present accentuation.⁵ His inclusion of 3 for his second edition would have only reinforced this reading as it too also accents ιαται as ἰᾶται.⁶ Although Erasmus’ accentuation ιαται of may be excused due to the limited number of manuscripts he employed, there may actually be enough evidence, both externally and internally, to suggest that a better reading of ιαται in Mark 5:29 is as a present tense-form and not the perfect.

The form ιαται occurs only twice in the NT, once in Mark 5:29 and once in Acts 9:34. While Mark 5:29 is accented as a perfect, Acts 9:34 is accented as a present tense-form. Bruce Metzger, drawing on a short article by Henry Cadbury, suggested that the scribe of Codex Vaticanus intended for the form in Acts 9:34 to be understood as a perfect as well: “The scribe of codex Vaticanus undoubtedly took the form to be the perfect tense, for he wrote it ΕΙΑΤΑΙ, as he did also at Mark 5:29 where there is no question that the perfect tense is intended.”⁷ There are two problems with this inference. First, it is now well-known that the Synoptics in Vaticanus have frequent itacistic changes, i.e. iota (ι) has frequently been replaced with epsilon-iota (ει).⁸ The extra epsilon therefore contributes nothing to the meaning of the verb other than as an orthographic habit of Vaticanus’ original scribe. Secondly, and more important for our purpose, neither Cadbury nor Metzger question that ιαται in Mark 5:29 is a perfect; both merely assume it as their basis for reading ιαται in Acts 9:34 as a perfect.

Rather than presenting ιαται in Mark 5:29 as a perfect tense-form, Vaticanus actually provides a witness of a scribal reader who understood ιαται to be a present tense-form. A corrector sometime in the 9–11th centuries went through Vaticanus and overwrote the letters, leaving out those s/he thought were erroneous, while also adding accents.⁹ The corrector did not overwrite the epsilon at Mark 5:29, recognising it as an erroneous addition. Importantly for us, the corrector also added a circumflex over the first alpha in ιαται, clarifying that it should be understood as the present tense-form (ἰᾶται).¹⁰

As the minuscules used by Erasmus testify, Vaticanus is not alone. The 9th century minuscule 33 (δ 48, Codex Colbertinus 2844)—also known as the “queen of the cursives”—now stored at the Bibliothèque nationale in Paris also has a circumflex over the alpha, indicating a present tense-form. This accentuation is further present in other 9th century minuscules, such as minuscules 461 and 1500 and continues through 10th century witnesses to the 13th century (e.g. 34 23 26 39 1 3 4 5).¹¹ These are by far the minority. Nevertheless, the presence of present tense-form readings alongside perfect readings indicate that some medieval scribes understood the present as a legitimate option for understanding ιαται in Mark 5:29.

There is further evidence that points toward the present tense-form as the intended meaning. Other than the instance in Mark 5:29 the only other time the perfect indicative third person singular of ἰάομαι occurs in extant Greek literature before the 12th century is in Origen’s explanation of Jer 3:22 OG (Hom. Jer. 5.1.2, PG 13.296c): ἕκαστος εἴποι ἄν, εἰ καὶ νῦν κεκαθάρισται καὶ ἴαται ἀπὸ τῶν συντριμμάτων … (“Each, if now having been cleansed, then healed from the afflictions, would say …”).¹² The verbal pairing with a morphologically explicit perfect verb form (κεκαθάρισται) makes clear that the perfect is intended in this context. Other than this instance, however, in the vast majority of other extant places that we know ιαται occurs (including Acts 9:34), the verb is understood as a present. Indeed, its rare appearance in our literature may only be circumstantial, but in the context of Mark there is a stylistic connection that strongly suggests the present tense-form was intended over the perfect.

3 Mark’s Use of the Present Tense-Form in Indirect Internal Discourse

It is not clear to me why interpreters in the past, represented by Metzger’s comment above, argue that the form of ιαται must have been in the perfect, other than to emphasise the completed action of the healing.¹³ It goes without saying that if the person is healed it has lasting results. Why then is it necessary for the perfect to demonstrate this verbal aspect? This focus on what is really a redundant feature of ιαται as a perfect tense-form diverts attention away from the way Mark uses the present tense in indirect discourse and how ιαται as a present tense-form in Mark 5:29 fits this pattern.

A stylistic tendency of Mark is to introduce indirect internal discourse by using a particular pattern: (a) the use of a verb of knowledge or cognition (e.g. ἐπιγινώσκω, οἴδα, νοέω, γινώσκω) followed by (b) ὅτι as a marker of indirect discourse, then (c) a clause whose main verb usually utilises a present tense-form (a “historical present”). To my knowledge, there are only two exceptions to the pattern displayed here: Mark 12:12, where the verb is an aorist, and Mark 15:10 with the pluperfect. Nevertheless, in the vast majority of instances the verb is in the present:

In particular, the parallels between Mark 5:29 and Mark 2:8 are striking:

Both passages share additional features with one another, such as the inclusion of the εὐθύς, the aorist participial form of a verb of knowledge, and the locus of where that knowledge occurs (in his spirit/in the body). In light of these close linguistic parallels with Mark 2:8 it makes sense to follow Mark’s stylistic pattern and to have the ιαται in Mark 5:29b be understood as a present not a perfect tense-form: ἔγνω τῷ σώματι ὅτι ἰᾶται ἀπὸ τῆς μάστιγος.

Our reading of the ιαται as a present tense-form is confirmed when we look at some of the early versional evidence for Mark 5:29. Unfortunately neither the early Syriac or Coptic versions are useful for our argument.¹⁴ Some of the witnesses of Old Latin translations of the Greek text render ιαται in Mark 5:29 with a corresponding present tense form, most notably Codex Vercellensis (a), Codex Veronensis (b), and Codex Vindobonensis (i), the first of which has been dated to the second half of the fourth century and the latter two to the end of the fifth century CE. In a b and i ὅτι ιαται has been rendered as quia sanata est (“that she is healed”).¹⁵ These three Old Latin witnesses also translate the present tense-forms of the Greek in the other passages examined in the pattern above (see Table 1).

View Full Size

A comparison of relevant verbs in Mark between NA²⁸, Old Latin, and Vulgate

Citation: Novum Testamentum 63, 3 (2021) ; 10.1163/15685365-12341699

• Download Figure
• Download figure as PowerPoint slide

As Hugh Houghton has cautioned, there is not a direct correspondence between Latin and Greek tenses and that “the tense of verbs often fluctuates in the Latin tradition regardless of the Greek Vorlage.”¹⁶ Nevertheless, the consistency with which the tenses of the passages above in a b and i match to the corresponding Greek tense-forms in Mark suggests that their translation is not by chance.

The one passage in these Old Latin witnesses that does not reflect the present tense is Mark 2:8; all three (a b i) in addition to the Vulgate render διαλογίζονται with cogitarent (third-person plural imperfect active subjunctive). Why would the present tense of the Greek be preserved so well elsewhere but not here in Mark 2:8? One possibility is that perhaps the Greek text used by translators had the verb διαλογίζειν in the imperfect (διελόγιζον) rather than the present. However, a limited analysis of all majuscules and early minuscules with Mark 2:8 present shows no minor variant of διαλογίζονται, let alone in the imperfect.

There is another explanation simply rooted in Greek original itself. The Latin witnesses a b and i show a tendency to smooth out the awkward presentation of the action in Mark 2:8. In the initial Greek text, the four main verbs skew the temporality of the action: καὶ εὐθὺς ἐπιγνοὺς ὁ Ἰησοῦς τῷ πνεύματι αὐτοῦ ὅτι οὕτως διαλογίζονται ἐν ἑαυτοῖς λέγει αὐτοῖς· Τί ταῦτα διαλογίζεσθε ἐν ταῖς καρδίαις ὑμῶν; (“And immediately Jesus, knowing in his spirit that they are considering this in themselves, is saying to them, ‘Why are you considering these things in your hearts?’”). The Latin of a clearly tries to smooth out the action: et cum cognovisset (third-person singular pluperfect active subjunctive of cognoscō) Iesus in spiritu, quod cogitarent intra se, dixit (third-person singular perfect active indicative of dicō) illis, quid haec cogitates (second-person plural present active indicative of cogitō) in cordibus vestris (“And when Jesus had known in the spirit what they were thinking to themselves, he said to them, ‘What is this you are thinking in your heart?’”). The other witnesses, b and i, follow this closely exception for the beginning of the verse which reads Quo cognito (masculine singular ablative perfect passive participle of cognoscō) instead of et cum cognovisset. It is not unreasonable then to assume that the present tense is not reflected in this one instance because the syntax of Mark 2:8 was updated for clarity by the Old Latin witnesses.

This versional evidence along with the rarity of the perfect in our extant literature, as well as the various manuscripts which attest to readings of the present, leans in favour that the present tense-form reading fits Mark’s literary context best. We should no longer read ιαται as ἴαται but as ἰᾶται, a present not a perfect tense-form.

4 Grammatical Subject, Agency, and the Present Tense-Form of ιαται

Understanding of the Greek verbal voice has undergone much clarification over the past century. The shift from an understanding of Greek as an active/passive voice system to an active/middle voice system has altered how we think about what is often taught as vocally sealed morphological systems (e.g. -(θ)η type verbs are always passive; so-called “deponent” verbs are “active in meaning but passive in form”).¹⁷ Ἰάομαι is what used to be called a “deponent” verb; its present tense-forms have been labelled morphologically middle/passive. Linguists have come to understand the terms “active,” “middle,” and “passive” as “waypoints along a continuum of descending transitivity” rather than impermeable categories.¹⁸ While ἰᾶται might be a middle tense-form, and a passive middle meaning is possible, it does not necessarily mean that in the case of Mark 5:29, the voice of the verb should be taken as such (i.e. “she is/was healed”). Our understanding of how the middle voice of ἰᾶται functions depends on literary context and lexical semantics.

Reading ιαται as present tense-form is significant because, to my knowledge, when we examine the lexical semantics of ἰᾶται in other literature, the grammatical subject of ἰᾶται is always the active agent of the action of healing, with only one exception (to be addressed below). When we look at the way that ἰᾶται is used in ancient Greek sources, its lexical semantics display two senses: active and passive middle. In the vast majority of examples, including the only other instance of ἰᾶται in the NT, the verb functions with an active voice whose grammatical subject is the agent doing the action of the verb, for example:

In all of these examples the grammatical subject of ἰᾶται is the agent who heals the object. In terms of its lexical semantics, the active voice appears as the default sense of ἰᾶται. This creates a problem with the conventional translation of 5:29, as the verb is usually rendered in a passive voice (“was healed”) with the woman as subject (“she”). One may be tempted, given the active role the woman plays in her whole healing, to suggest that the ἰᾶται points to the woman in Mark 5:29 as the active agent in her own healing (e.g. “She healed herself from the affliction”).¹⁹ However, if the verb was meant to be understood as reflexive we might expect some a reflexive pronoun (e.g. ἑαυτήν/αὐτήν) to accompany the verb as we do in what appears to be the only reflexive instance of ἰᾶται found in an epitome of Aristotle’s Historia animalium by Aristophanes of Byzantium (2.408): Ἀλώπηξ βέλει τρωθεῖσα τῷ δακρύῳ τῆς πίτυος ἑαυτὴν ἰᾶται (“A fox wounded by an arrow heals itself with the sap from the pine”).²⁰

Alternatively, perhaps the subject of ἰᾶται is Jesus himself, since the power that does cause the woman to become well originates with him. Although the shift from her perspective (ἔγνω τῷ σώματι ὅτι) to his perspective (ἴαται ἀπὸ τῆς μάστιγος) may appear jarring, this is not outside Mark’s style of conveying indirect internal discourse. Similarly, in Mark 2:8, the text portrays Jesus self-knowledge from his own perspective (ἐπιγνοὺς ὁ Ἰησοῦς τῷ πνεύματι αὐτοῦ ὅτι) only to then switch to the perspective of the scribes (οὕτως διαλογίζονται ἐν ἑαυτοῖς). However, if this were the case, then like the reflexive middle, we would expect an explicit textual object, that Jesus healed her (αὐτήν) or it (αὐτό), that is her body, from the affliction. Also, given the ambiguity produced by having two third person singular verbs one after the other (ἔγνω … ἰᾶται), the first of which is clearly the woman, a clarifying pronoun or proper noun (e.g. Ἰησοῦς) would be expected if the second verb had a different subject.²¹

There is, however, a single extant use of ἰάομαι in a passive middle sense found in Lev 14:3 OG.²² However, the context of Lev 14:3—especially in relation to its Hebrew Vorlage—makes a comparison with Mark 5:29 problematic:

Interpreters of the Hebrew have dealt with the awkwardness of the syntax here by translating the passage with subject and object reversed: “The scale-diseased person has been healed of scale disease.”²³ The OG translator(s) of Lev 14:3 have rigidly rendered the awkwardness of the Hebrew into the Greek, from the reversal of the expected subject and object to the odd prepositional phrase beginning with ἀπό. It is obvious that ἰᾶται is meant to be read as a passive middle reflecting the passive Niphal perfect form of רפא in Lev 14:3. The syntax of the verse, however, while sharing both the verb ἰᾶται and prepositional phrase with ἀπό is not the same as Mark 5:29. In the latter, the syntax follows the logical subject/object placement, with the woman as the grammatical subject of the verb and the affliction as the condition from which she is healed (cf. Luke 6:18). Although not precisely the same, this is nevertheless the only comparative example outside of the text that can explain how ἰᾶται functions semantically as a passive middle in Mark 5:29.

Turning to the material evidence, an early witness of Mark 5:29, the 5th century Codex Washingtonianus (032), understood ἰᾶται to be functioning passively. Either its scribe or its exemplar has replaced ΙΑΤΑΙ with ΕΙΑΘΗ, which—taking into account itaticisation—indicates a third person singular middle/passive indicative form of ἰάομαι (“she was healed”). Later minuscules also reflect this reading, though without the extra epsilon, for example ιαθη in minuscules 28 (11th century) and 6 (13th century).²⁴

Looking to the grammatical subject of the verb may help us clarify the verb’s semantics in this instance. What then is the grammatical subject of ἰᾶται? We have already ruled out above that Jesus is the grammatical subject. Even though the woman may not be the agent of healing, she may nevertheless be the grammatical subject of the verb if it functions as a passive middle, which is the traditional reading of the text. However, in light of the fact that it is indirect discourse and that this is something she is presented as thinking to herself, it is unlikely that she would refer to herself in the third person; one would expect indirect discourse and the verb ἰάομαι to be in a first person morphology (ἰῶμαι).

On the other hand, an even simpler answer is that the grammatical subject of ἰᾶται is the woman’s body, her σῶμα. There are various pieces of evidence that point to this reading. First, a reference to her body (τῷ σώματι) immediately precedes the indirect speech. Secondly, an implicit reference to her body fits the third person singular form of the ἰάομαι in the context of Mark’s indirect discourse. It makes good sense that the woman might be portrayed as thinking to herself, “It is healed from the affliction.” On this, it may be best to understand the grammatical subject to be the woman’s body.

Either way however, the verb ἰᾶται must be functioning as a passive middle. In light of this analysis, Mark 5:29 should thus be translated: “Immediately, the flow of her blood dried up and she knew in the body, ‘It is healed from the affliction’,” or more smoothly, “Immediately, the flow of her blood dried up and she knew in the body that it was healed from the affliction.”

5 Conclusion

This short article has argued that there is sufficient evidence for reading ιαται in Mark 5:29 as the present tense-form ἰᾶται rather than the perfect tense-form ἴαται. The accentuation in ancient Greek witnesses presents both the present and the perfect as viable interpretations. While, the perfect ἴαται occurs dramatically less than the present tense-form, reading ιαται in Mark 5:29 as ἰᾶται fits Mark’s use of present tense-form verbs for indirect internal discourse. Old Latin versions of Mark confirm this reading. While the majority of times ἰᾶται appears in ancient Greek literature it undeniably functions as an active verb, the evidence of Lev 14:3 along with the grammatical subject of the verb—the woman’s body (“it”)—reinforce the traditional understanding of the voice of the verb as a passive middle. This article revives an accentuation of ιαται present in a number of medieval minuscules that has been neglected in critical editions of the Greek New Testament since Erasmus.

One should be cautious to merely accept the accentuation of critical editions ex facie, especially given our present digital access to so many of the manuscript witnesses, e.g. via the INTF’s New Testament Virtual Manuscript Room (NTVMR).²⁵ To an uninformed reader, critical editions of the NT can present artificial discrepancies.²⁶ The way our critical editions of the NT present artificially the text of our earliest witnesses is evident not only in what they exclude (e.g. scriptio continua) but what features they do include, such as accentuation. In this way NT critical editions can function much like the Masoretic vocalisation of the Hebrew Bible. The Masoretic vocalisation is one way of understanding the unvocalised text. But, as ancient translations of the HB show, such as the OG, the unvocalised text could be understood in many different ways. In the same way that it has been argued that the Greek translation of the Pentateuch argues against a monolithic fixed community vocalisation of the unvocalised Hebrew text, the varying ways in which ιαται is rendered in the ancient versions, as well as its variable accentuation among medieval witnesses, shows just how accentuation of the Greek NT is sometimes much less uniform than our critical editions depict.²⁷

In order to provide more manuscript transparency, critical editions of the NT could leave words unaccented and show different accenting possibilities. For example, ιαται in Mark 5:29 might be accompanied with an added apparatus showing the various ways it is attested and their relevant witnesses:

ἴαται 07 09 013 028 037 399 566 892 1424 2811 l 63 l 848 l 849 ¦ ἰᾶται B^s 1 3 4 5 23 26 34 39 etc.

Indeed, in preparations for the ECM of the Apocalypse, scholars are taking into account variant accentuation in the manuscripts, hopefully to be included in the apparatus itself.²⁸ Providing this resource in critical editions will allow users to make informed judgments for themselves on variant accentuation without having to laboriously scour through all the extant material evidence.